The present invention relates to an automatic performance apparatus and, more particularly, to an improvement in a performance information input/memory section
A conventional example of an automatic performance apparatus stores performance information such as a pitch and a tone duration based on an operation of an input operating member such as a key in a memory and generates a musical tone on the basis of the performance information stored in the memory.
In some automatic performance apparatuses of this type, a read-back switch is provided for correcting inputted information so that each time the read-back switch is pushed on, the performance information is read out for one tone after another and converted into sounds in an order opposite to the inputted order, thereby searching performance information to be corrected. This searched performance information is corrected by, e.g., rewriting into the memory.
However, according to a conventional performance information correcting means as described above, if, for example, a plurality of tones having the same pitch continue, it is often difficult to determine performance information corresponding to a tone to be corrected. Therefore, it is difficult to accurately perform a correction operation.
In addition, a generated tone must be aurally perceived and examined each time the read-back switch is operated, resulting in a troublesome searching operation.